I Recorded Sound At A Wedding... To MP3... 64K...

[David W. Richardson]

I didn't think so but thought I would ask!

Here is a link to a sample of the recording: https://www.dropbox.com/s/gvatsgxo3p9o6bn/TASCAM_0253_R.mp3?dl=0

Thanks

It sounds like you'd already applied at least some noise reduction on that clip. It would help if we heard the original clip.

You don't specify in what way it sounded "bad", so how can we know if it's due to being recorded in MP3 format, or maybe has more to do with bad mic placement or the acoustics of the location?

 

[Hunter Starnes]

It sounds like you'd already applied at least some noise reduction on that clip. It would help if we heard the original clip.

You don't specify in what way it sounded "bad", so how can we know if it's due to being recorded in MP3 format, or maybe has more to do with bad mic placement or the acoustics of the location?

No, this was from the original file without any work done to it. This is just a snippet from the recording. I am not sure why I re-encoded it to mp3. I should have uploaded the original file which is VERY small in size.

 

[AtticusLake]

I actually question your assessment of what ProTools is doing. One of the fights they have constantly had with Apple is that they tend to work around the OS. Properly written there shouldn't need to be a huge delay. But you know I'm not a real software developer so?

I actually am a software developer (35 years plus)... I don't know specifically what ProTools is doing internally, but it's actually hard to make the OS ACTUALLY write something to the disk. And when you do there's a huge performance hit, which is why it's rarely done - at least not on every write, maybe periodically.

Then, getting it to PHYSICALLY read the data back off the disk -- as opposed to the filesystem cache, or the driver's cache, let alone the disk drive's internal cache -- I don't klnow if that can even be done. Maybe, but by killing off all the cacheing, there would again be a massive performance hit.

Flash memory writes slow so it makes sense that you can't get decent off card monitoring, not really comparable with a high speed Hard drive.

Also very true.

 

[Rick R]

I don't know about SloTools, but in SCS software (Vegas, Sound Forge, ect.), when recording or editing, the data is 32 bit (floating point) in the RAM and/or page file, and not saved to the drive(s) until the user initiates 'save'. AFAIK.. I'm not a software developer.

 

[Noiz2]

I'll bow to your expertise on software development. I do think that "massive" and "Huge" may be relative to computer speeds and not "real life" speeds. As in if a portable recorder can write to and then read off a flash card with a few second lag I would assume that a hard drive would be a small fraction of that time.

that said it may well bee too big a lag for normal use.

Dave,

Yes in theory that would be a plus. The same is true with the old HHb PortaDAT which could go into "record" with out actually recording if you somehow got past the end of what had been recorded and started recording. That "feature" nailed a number of people I know including me. I assume it would have anyway. Nobody was ever quite sure if it didn't record or just did so in a unplayable fashion, ie lost the sync. Whatever it did you ended up with what appeared to be blank tape. While "recording" everything looked and acted normal, TC counted meters bounced etc.

It happened mostly to FX recordists because we were constantly popping tapes out to transfer something and then using it to record again. THe HHb wouldn't let you go past the end of the recorded section but a lot of the rack DATs we used for transfer would so you had to be really careful that you rewound and then forwarded to the end IN the HHb.

But even with that at most folks would switch to off tape monitoring for maybe a touch after putting a tape back in and then switch back because it's such a PITA having the lag. With the Marantz you are a couple of words behind the action, that is HUGE. Just think of being five seconds behind the rest of the crew on every "cut". If you are doing any mixing you are five seconds off on every move. So if you balance those off against the (hopefully) very rare occasions where you just screw up it's just not worth it for most people.

In this case it probably wouldn't have helped either. I mean what are his options? Stand up and tell everyone to stay put while he tries to figure out why it sounds bad? Even if he catches it right away he has to go into menus to change the settings and then start recording again. So he would either have to get them to start over or lose a hunk of the ceremony altogether. So in reality IF he figured it out all that would have happened is that he would know sooner that the recording was going to be an issue.

Of course he could have done a short equipment test before the ceremony, I always record a bit to make sure everything is working before I need them working. But in that case he should have caught the error either way.

 

[Noiz2]

I don't know about SloTools, but in SCS software (Vegas, Sound Forge, ect.), when recording or editing, the data is 32 bit (floating point) in the RAM and/or page file, and not saved to the drive(s) until the user initiates 'save'. AFAIK.. I'm not a software developer.

Seriously? So your recording is totally at risk till you can do a save command? That is scary. Reminds me of the times I lost long takes on MiniDisk because it kept the file open till it stopped recording which meant if the batteries died or it had a glitch your file was gone. I bet you have a nice big UPS on that computer ;~)

My first thought was that it would really limit recording time... Then my brain caught up to 2016! Sometimes I flash back to 650Megs being a lot of memory.

 

[Noiz2]

I obviously have too much time today.

But from a practical standpoint the drive/ card is just a bit bucket. So really if you can monitor that digital stream there isn't a lot of point monitoring off the drive. The drive isn't storing the "recording" it's storing a string of digits and file systems have check built in that verify the integrity of the data.

Going back to the OP IF the device monitored the stream going to the card then that would have been the MP3 data, so it should have sounded like what he ended up with.

So that would be a very interesting question. Where exactly are the various devices monitoring from? I assume most are monitoring from the input analog stage. But some of the Tascam's are doing at least some of the gain etc after the signal has been digitized and since that shows up in the headphones it must be monitoring post A/D.

 

[Rick R]

Scott, I never actually tested that scenario by 'pulling the plug' whilst recording, nor do I record directly to PC much anyway, except digitizing a LP, cassette or other analog source material. Though.. if editing w/ Vegas and/or Sound Forge, if the system crashes, there are 'temp' recovery files to revert to, including the last edit or process prior to system crash.
Sorry for getting way off-topic.

 

[davefisk]

Of course he could have done a short equipment test before the ceremony, I always record a bit to make sure everything is working before I need them working. But in that case he should have caught the error either way.

Yes. ALWAYS prep gear before a gig to make sure everything is set where it is supposed to be and works properly. Maybe that was done in this specific scenario and something weird happened anyway. I've always personally prepped my own gear before every gig, and there have been a few times where I get to the gig and something acts up for some weird reason, but I've been lucky that I've caught it and been able to solve the problem quickly.

I'm always a big fan of using the right tool for the right job, and I've never been a fan of dropping a recorder into someone's pocket and walking away. I've read enough posts on here to see it's burned people a lot. I think at minimum a wireless system should be used. At most, a recording wireless. If the actual recorder is with you, and not on someone across the room in the church, you can fix the problem within seconds and nobody is the wiser.

 

[Noiz2]

Scott, I never actually tested that scenario by 'pulling the plug' whilst recording, nor do I record directly to PC much anyway, except digitizing a LP, cassette or other analog source material. Though.. if editing w/ Vegas and/or Sound Forge, if the system crashes, there are 'temp' recovery files to revert to, including the last edit or process prior to system crash.
Sorry for getting way off-topic.

It's OK. That sounds like your covered. I usually a run till dry test on portable gear. Most things shut down properly and give you warning. The MD was an unfortunate exception.

I actually record to the system fairly often, not like a music recording studio but fairly often.

Early versions of PT would corupt session files if the plug got pulled. They still might in some cases, but with auto backup you only loose a few minutes of work at worst.

I can't say I have pulled the plug on a PT recording though. I did with BoomRecorder and lost nothing.

 

[Noiz2]

;~) I had the power supply go out on a 24X16 Neotek board between sound check and house open. Had to rough patch a 16X4 channel board in and run the show blind because to make the connections it had to sit on the floor of the booth and of course all levels were educated guesses with the switch. That was fun! Nothing like walking in and seeing every LED on the board lit up and every meter pegged with nothing running just as the pre show should be starting.